Prev: proc mdc
Next: Predicted probabilities in GLIMMIX
From: Shawn Haskell on 10 Dec 2009 12:47
On Dec 9, 1:48 pm, dy...(a)CHAPINHALL.ORG (Duck-Hye Yang) wrote:
> Dear SAS-L,
>
> I have a hard time in interpreting interaction term in a proportional cox model.
> Basically I have two variables, group (treatment=1; control=0) and period (post=1; pre=0) and an interaction term consisting of those two (grouppost).
>
> The output I got is as follow:
>                               beta       S.E               Chi-Square    Pr>ChiSq       Hazard Ratio
> group             1       0.20744       0.02129       94.9269        <.0001             1.231
> post               1        0.01789       0.02106        0.7219        0.3955              1.018
> grouppost      1      -0.06118       0.02924        4.3785        0.0364               0.941
>
> At Pre, the beta of the group is 0.20744. At post, what is the beta of the group? Is it multiplicative so that it is (0.20744 * -0.06118) or additive (0.20744 + -0.06118) or anything else?
>
> Thank you very much in advance.
>
> Duckhye Yang

The interaction just means the effect of Group depends on Period, or
vice-versa. The best way to see what it means is to plot your data
(Kaplan-Meier) stratified by your 2 groups or get Cox PH model
predictions - you'll see what's going on. SH
 | 
Pages: 1
Prev: proc mdc
Next: Predicted probabilities in GLIMMIX